P300 Detection Research Articles

Robust extraction of P300 using constrained ICA for BCI applications

P300 is a positive event-related potential used by P300-brain computer interfaces (BCIs) as a means of communication with external devices. One of the main requirements of any P300-based BCI is accuracy and time efficiency for P300 extraction and detection. Among many attempted techniques, independent component analysis (ICA) is currently the most popular P300 extraction technique. However, since ICA extracts multiple independent components (ICs), its use requires careful selection of ICs containing P300 responses, which limits the number of channels available for computational efficiency. Here, we propose a novel procedure for P300 extraction and detection using constrained independent component analysis (cICA) through which we can directly extract only P300-relevant ICs. We tested our procedure on two standard datasets collected from healthy and disabled subjects. We tested our procedure on these datasets and compared their respective performances with a conventional ICA-based procedure. Our results demonstrate that the cICA-based method was more reliable and less computationally expensive, and was able to achieve 97 and 91.6% accuracy in P300 detection from healthy and disabled subjects, respectively. In recognizing target characters and images, our approach achieved 95 and 90.25% success in healthy and disabled individuals, whereas use of ICA only achieved 83 and 72.25%, respectively. In terms of information transfer rate, our results indicate that the ICA-based procedure optimally performs with a limited number of channels (typically three), but with a higher number of available channels (>3), its performance deteriorates and the cICA-based one performs better.

P300 in neglect

ObjectiveAttentional deficits have been thought to underlie hemispatial neglect. The present study investigates this using the P300 component of the event-related potential, which is assumed to reflect the allocation of attention. MethodsWe recorded the P300 and oddball detection performance from patients with neglect. Infrequent targets appeared either in the affected contralesional or the unaffected ipsilesional visual hemifield. ResultsContralesional targets produced weaker average P300 responses than ipsilesional ones, but stronger responses than when only distractors were presented. In the affected hemifield, the difference in P300 amplitude between identified and missed targets was correlated with the miss rate. The number of missed contralesional targets for a given participant was positively correlated with the time required for completion of standard neglect tests. ConclusionsThe P300 not only reflects a general attentional impairment in neglect, but also varies as a function of the participants’ awareness of individual stimuli. SignificanceThe findings reflect an important link between attentional dysfunction and impaired stimulus processing in neglect and extinction.

Adaptive training session for a P300 speller brain–computer interface

With a brain–computer interface (BCI), it is nowadays possible to achieve a direct pathway between the brain and computers thanks to the analysis of some particular brain activities. The detection of even-related potentials, like the P300 in the oddball paradigm exploited in P300-speller, provides a way to create BCIs by assigning several detected ERP to a command. Due to the noise present in the electroencephalographic signal, the detection of an ERP and its different components requires efficient signal processing and machine learning techniques. As a consequence, a calibration session is needed for training the models, which can be a drawback if its duration is too long. Although the model depends on the subject, the goal is to provide a reliable model for the P300 detection over time. In this study, we propose a new method to evaluate the optimal number of symbols (i.e. the number of ERP that shall be detected given a determined target probability) that should be spelt during the calibration process. The goal is to provide a usable system with a minimum calibration duration and such that it can automatically switch between the training and online sessions. The method allows to adaptively adjust the number of training symbols to each subject. The evaluation has been tested on data recorded on 20 healthy subjects. This procedure lets drastically reduced the calibration session: height symbols during the training session reach an initialized system with an average accuracy of 80% after five epochs.

Convolutional Neural Networks for P300 Detection with Application to Brain-Computer Interfaces

A Brain-Computer Interface (BCI) is a specific type of human-computer interface that enables the direct communication between human and computers by analyzing brain measurements. Oddball paradigms are used in BCI to generate event-related potentials (ERPs), like the P300 wave, on targets selected by the user. A P300 speller is based on this principle, where the detection of P300 waves allows the user to write characters. The P300 speller is composed of two classification problems. The first classification is to detect the presence of a P300 in the electroencephalogram (EEG). The second one corresponds to the combination of different P300 responses for determining the right character to spell. A new method for the detection of P300 waves is presented. This model is based on a convolutional neural network (CNN). The topology of the network is adapted to the detection of P300 waves in the time domain. Seven classifiers based on the CNN are proposed: four single classifiers with different features set and three multiclassifiers. These models are tested and compared on the Data set II of the third BCI competition. The best result is obtained with a multiclassifier solution with a recognition rate of 95.5 percent, without channel selection before the classification. The proposed approach provides also a new way for analyzing brain activities due to the receptive field of the CNN models.

P300 Detection Algorithm Based on Fisher Distance

With the aim to improve the divisibility of the features extracted by wavelet transformation in P300 detection, we researched the P300 frequency domain of event related potentials and the influence of mother wavelet selection towards the divisibility of extracted features, and then a novel P300 feature extraction method based on wavelet transform and Fisher distance. This can select features dynamically for a particular subject and thereby overcome the drawbacks of no systematic feature selection method during traditional P300 feature extraction based on wavelet transform. In this paper, both the BCI Competition 2003 and the BCI Competition 2005 data sets of P300 were used for validation, the experiment results showed that the proposed method can increase the divisibility by 121.8% of the features extracted by wavelet transformation, and the classification results showed that the proposed method can increase the classification accuracy by 1.2% while reduce 73.5% of the classification time. At the same time, integration of multi-domain algorithm is proposed based on the research of EEG feature extraction algorithm, and can be utilized in EEG preprocessing and feature extraction, even classification.

P300 Detection Algorithm Based on Fisher Distance

P300 Detection Algorithm Based on Fisher Distance

Detection and Analysis of Event-Related Potential P300 in EEG by 4-Stimulus Oddball Paradigm

P300 component of the event-related potential(ERP) has been studied for theoretical, empirical, and clinical applications. Nowadays the 1-, 2-, and 3-stimulus oddball paradigms are used for eliciting P300 component of EEG in the auditory stimulus experiments. In this paper, we used a method to add one more stimulus to the 3-stimulus auditory paradigm. The adding stimulus has not the same volume but the same tone of the target stimulus. The 4-stimulus oddball paradigm to use two targets is used to elicit the P300 event-related potentials. In 4-stimulus oddball paradigm, an infrequent non-target (p=0.10) is presented in addition to two infrequent targets (p=0.10) and a frequent standard (p=0.70). Two target stimuli elicited a P300 component with a parietal maximum distribution. The amplitude of the P300 in target 2 was larger than that in target 1 and the latency of the P300 in target 2 was longer than that in target 1. The P300 component due to target 2 stimuli was larger than that due to target 1 stimuli. The experimental results approve that the 4-stimulus oddball paradigm can elicit P300 component clearly. The results are compared with the results of the traditional oddball paradigm.

Online detection of P300 and error potentials in a BCI speller.

Error potentials (ErrPs), that is, alterations of the EEG traces related to the subject perception of erroneous responses, have been suggested to be an elegant way to recognize misinterpreted commands in brain-computer interface (BCI) systems. We implemented a P300-based BCI speller that uses a genetic algorithm (GA) to detect P300s, and added an automatic error-correction system (ECS) based on the single-sweep detection of ErrPs. The developed system was tested on-line on three subjects and here we report preliminary results. In two out of three subjects, the GA provided a good performance in detecting P300 (90% and 60% accuracy with 5 repetitions), and it was possible to detect ErrP with an accuracy (roughly 60%) well above the chance level. In our knowledge, this is the first time that ErrP detection is performed on-line in a P300-based BCI. Preliminary results are encouraging, but further refinements are needed to improve performances.

The Role of Intention to Conceal in the P300-based Concealed Information Test

The present study examined whether intention to conceal knowledge affects P300 amplitude and detection accuracy in the concealed information test. Eighteen university students were told to choose one card from five and to hide it. In the conceal condition, participants made an effort to leave their chosen card undetected by suppressing their brain response to it. In the transmit condition, they attempted to inform the experimenter of the chosen card by enhancing brain response to it. In the no secret condition, participants showed the chosen card to the experimenter beforehand and lost their motivation to conceal it. The difference in P300 amplitude between the chosen and unchosen cards was significant only in the conceal and transmit conditions. The results suggest that a larger P300 amplitude for the chosen card was not due to a deception-specific process but rather to increased significance of the item caused by additional processing.

A region-based P300 speller for brain-computer interface

A brain-computer interface (BCI) is a system that conveys messages and commands directly from the human brain to a computer. The BCI system described in this work is based on the P300 wave. The P300 is a positive peak of an event-related potential (ERP) that occurs 300 ms after a stimulus. One of the best-known and most widely used P300 applications is the P300 speller designed by Farwell-Donchin in 1988. The Farwell-Donchin paradigm has been a benchmark for P300 BCIs. In this paradigm, a 6 X 6 matrix of letters and numbers is displayed, and the subject focuses on a target character while rows and columns of characters flash. Through detection of P300 for one row and one column, the target character can be identified. In this paper, it is shown that there is a human perceptual error in the Farwell-Donchin paradigm. To eliminate this error, a new region-based paradigm is presented. Using experimental results, it is shown that the new paradigm has several advantages over the Farwell-Donchin paradigm and achieves better accuracy.

EVALUATION OF THE HIDDEN MARKOV MODEL FOR DETECTION OF P300 IN EEG SIGNALS

Introduction: Evoked potentials arisen by stimulating the brain can be utilized as a communication tool between humans and machines. Most brain-computer interface (BCI) systems use the P300 component, which is an evoked potential. In this paper, we evaluate the use of the hidden Markov model (HMM) for detection of P300. Materials and Methods: The wavelet transforms, wavelet-enhanced independent component analysis (W- ICA), and HMM combined with a multi-layer perceptron (MLP) neural network were used for P300 detection in electroencephalogram (EEG) signals. The BCI2005 competition dataset was used for their evaluation. First, electrooculogram (EOG) artifacts in the EEG signals were removed using W-ICA. Then, background EEG noise was suppressed using a B-Spline wavelet transform. Finally, these signals were classified using the HMM. Results: We used accuracy, sensitivity, specificity, positive predictive value, and negative predictive value to evaluate the performance of the proposed algorithm. The primary results in this research show that the HMM can perform much better using an auxiliary classifier. To this end, an MLP neural network was used to select the classes based on the outputs of the HMM models. The classification rates obtained for 15 and 5 times averaged test signals were 81.6% and 50.7% respectively. Discussion and Conclusion: Based on the obtained results, we may conclude that the HMM can be used for online P300 detection.

A new approach for EEG feature extraction in P300-based lie detection

P300-based Guilty Knowledge Test (GKT) has been suggested as an alternative approach for conventional polygraphy. The purpose of this study was to extend a previously introduced pattern recognition method for the ERP assessment in this application. This extension was done by the further extending the feature set and also the employing a method for the selection of optimal features. For the evaluation of the method, several subjects went through the designed GKT paradigm and their respective brain signals were recorded. Next, a P300 detection approach based on some features and a statistical classifier was implemented. The optimal feature set was selected using a genetic algorithm from a primary feature set including some morphological, frequency and wavelet features and was used for the classification of the data. The rates of correct detection in guilty and innocent subjects were 86%, which was better than other previously used methods.

Fusion of classic P300 detection methods’ inferences in a framework of fuzzy labels

Designing a reliable and accurate brain-computer interface (BCI) is one of the most challenging fields in biomedical signal processing. To achieve this goal, different methods have been adopted in different blocks of a typical BCI system (i.e., in preprocessing, feature extraction, feature classification and feature selection blocks). Since BCI's speed plays a crucial role in its success in real-life applications, using mathematically simple techniques with accurate and reliable performance can improve this aspect of BCI systems' design. In this paper, a new method is introduced, which combines information from different classic time series similarity measures, using a simple fuzzy fusion framework. This method is accurate and reliable in P300 (a positive event-related component occurring 300 ms after stimulus onset) detection. This framework is used to combine two computationally simple signal detection methods: "peak picking" and "template matching". Fusion takes place in the last step (decision-making step) by means of a fuzzy rule-base. Compared to similar works on electroencephalogram-based (EEG-based) BCI datasets, in spite of being computationally simple, this new technique's performance is comparable to very complicated methods, like support vector machines. This research indicates that, using both spatial and temporal information content of EEG trials (from all electrodes or a subset of them), even under a non-complicated mathematical framework can yield an accurate and powerful classification.

Single versus multiple probe blocks of P300-based concealed information tests for self-referring versus incidentally obtained information

The aim of the present study was to compare two protocols and two information types on detection of concealed knowledge. Four independent groups of subjects were run. Two were tested on probe stimuli of a self-referring (AUTO) nature and two were tested on incidentally acquired details of a mock crime (MOCK). Each pair of groups was run either in a one-probe (1PB) or multiple probe (6PB) block. In the single probe block, which was repeated three times with a different probe on each block, one probe item was randomly repeated multiple times in a Bernoulli series with frequently presented, meaningless or irrelevant items. There was also a rare target item designed to force attention to the stimulus screen. All stimulus types were of the same category within each block. In contrast, in the multiple probe (and category) block, rare probes, rare targets and frequent irrelevant items were repeatedly presented in a Bernoulli series within one block. Major results: There was a difference in task demand as measured by reaction time between the two protocols (the multiple probe protocol was more demanding), and a difference trend in P300 detection sensitivity between protocols for both information types combined in favor of the 1PB ( p < .07). With both protocols combined, there was a trend ( p < .07) favoring detection of familiar versus incidentally learned information. In terms of P300 amplitudes, both protocols showed the usual result that P300 to probes was greater than that to irrelevants. Also, as with detection rates, self-referring information was better detected in terms of Probe-Irrelevant P300 amplitude differences than mock crime information, regardless of protocol. There was no effect of time passage across the repeated blocks of the one-probe protocol. Methodological problems with the multiple probe protocol as utilized in most recent publications are discussed.

Detection of P300 in children with tic disorders

目的 了解抽动障碍患者事件相关电位P300的变化及意义.方法用日本光电公司提供的MEB-9104K诱发电位检测系统测定41例抽动障碍患者(病例组)P300并与22例正常儿童(对照组)进行比较.结果病例组P300潜伏期为(582.7±137.4)ms,波幅为(24.78±10.25)μV,对照组潜伏期为(501.2±126.0)ms,波幅为(23.15±10.00)μV.2组潜伏期均值比较,差异有显著性(P＜0.01);波幅均值比较,差异无显著性(P＞0.05).病例组各亚型之间P300潜伏期及波幅比较,均差异无显著性(P＞0.05).病例组中治疗组与未治疗组之间P300潜伏期及波幅比较,均差异无显著性(P＞0.05).结论抽动障碍患者存在认知功能缺陷,在治疗该类患者的过程中值得参考.抽动障碍各亚型之间认知功能无明显差别.氟哌啶醇、泰必利等药物治疗抽动障碍对认知功能无影响。

Comparison of the P300 and Other Workload Assessment Techniques in a Simulated Flight Task

Emerging cockpit technologies, such as graphical weather information systems, frequently shift perceptual demands from auditory to visual channels. Establishing methods of comparing the mental workload required by visual versus auditory displays within multitask environments is critical to maintaining aviation safety. Toward this end, the current investigation used a secondary task paradigm to compare physiological (i.e. peak amplitude and peak latency of the P300 ERP component), subjective and behavioral performance measures of mental workload required by a visual and auditory flight task. Participants flew a simulated flight path scenario while responding to either single or multiple auditory and text-based ATC commands and performing a visual or auditory detection task. Results indicate the P300 component, detection accuracy, response time and TLX ratings were sensitive to the presence or absence of the primary flight task. Accuracy and response time distinguished between single and multiple task command conditions. Results provide support for the suitability of this dual task technique for comparing displays of different modalities.

Automatic detection method of P300 waveform in the single sweep records by using a neural network

Event-related potentials (ERPs) are responses related to the recognition of certain stimuli. P300 is the most important positive component in the ERP and appears around 300 ms after the target stimulus in the oddball paradigm. In our previous work, we proposed a method for the automatic detection of the P300 waveform in single-sweep records by using a correlation technique. However, determination of the threshold values of the P300 waveform for the correlation study was not an easy task. In skirting this problem, we developed an automatic method of detecting a single-sweep P300 waveform by using an artificial neural network. We selected appropriate characteristic parameters of positive peaks as input signals for the input layer units, and the weights between the layers were determined by using the back-propagation algorithm. The neural network for P300 detection was obtained automatically, based on the data of ERPs obtained from 11 healthy males, and gave substantial accuracy for P300 detection. Furthermore, by using this neural network we clarified the way in which the P300 waveform is judged visually by each inspector.

Automatic detection of P300 in single sweep records of auditory event-related potential.

In the conventional recording of event-related potentials (ERPs) by the "oddball" paradigm, the subject may employ different strategies for processing one stimulus from the other even within the same session because of variable psychological state. Hence, "averaging" may not necessarily be suitable for recording ERP. We have therefore established a method for single trial recording of auditory ERPs by processing the EEG data through discrete Fourier transform followed by inverse discrete Fourier transform with the low pass filter of 8 Hz (digital filter). Each single sweep response of the processed data was visually evaluated as to whether it contained P300 or not by three investigators who were made completely blind with respect to the category of each response. As the result, a positive peak at a latency of 200-500 ms was recognized in the single sweep responses to target stimuli at a significantly greater frequency (40-60%) than in those to nontarget stimuli (0-18%). Then a new method for the automatic detection of P300 was established based on the data of 11 subjects belonging to the training group, by taking into account, in addition to the correlation coefficient analysis in which the template was computed specifically for each subject, other criteria with respect to the amplitude and waveforms, so that a good agreement with the results of visual inspection as to the presence or absence of P300 could be achieved for both target and nontarget responses. This automatic analysis method was successfully applied to five other subjects of the test group. The peak latency and amplitude of the detected positive peaks significantly varied among single target responses even within the same session. The reaction time significantly correlated positively with the peak latency of the detected positive peaks and negatively with the amplitude of those peaks, supporting the usefulness of the present method.

Brain Computer Interfaces for communication and control

Event Abstract Back to Event Brain Computer Interfaces for communication and control Fabio Babiloni1* 1 Imperial College London, United Kingdom Aim: The aim of this paper is to analyze whether the use of the cortical activity estimated from non invasive EEG recordings could be useful to detect mental states related to the imagination of limb movements and how this technology could be useful in clinical applications. Methods: Estimation of cortical activity was performed on high resolution EEG data related to the imagination of limb movements and detection of P300 gathered in a group of normal healthy subjects by using realistic head models. Cortical activity was estimated in Region of Interest associated with the subject's Brodmann areas by using depth-weighted minimum norm solutions. Comparisons between surface recorded EEG and the estimated cortical activity were performed. Subjects are asked to interact with different electronic devices of common use by using such BCI system. A restricted group of patients suffering by cerebral stroke were also involved in the BCI experiments. Motor imagery for stroke patients is used in the context of rehabilitation tasks. Results: With the proposed methodology healthy users are able to drive several mechanical and electronic devices by modulating their EEG activity related to motor imagery. BCI-based P300 is found a methodology easier to use for the stroke patients than motor imagery for driving devices. Rehabilitation tasks in patients involving motor imagery could be checked by using BCI system Conclusions: EEG recordings can be used to drive electronic and mechanical devices in healthy and in a group of patients suffering from stroke. A possible role of BCI system in rehabilitation tasks for certain class of stroke patients has been investigated. The presented approach could extend the possibility for a use of BCI system into the rehabilitation path for those patients. Conference: NeuroMath COST Action BM0601: Neurodynamic Insight into Functional Connectivity, Cognition, and Consciousness, Dubrovnik, Croatia, 27 Mar - 28 Mar, 2010. Presentation Type: Oral Presentation Topic: Talks Citation: Babiloni F (2010). Brain Computer Interfaces for communication and control. Front. Neurosci. Conference Abstract: NeuroMath COST Action BM0601: Neurodynamic Insight into Functional Connectivity, Cognition, and Consciousness. doi: 10.3389/conf.fnins.2010.05.00007 Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters. The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated. Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed. For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions. Received: 24 Jul 2010; Published Online: 24 Jul 2010. * Correspondence: Fabio Babiloni, Imperial College London, Paris, United Kingdom, nemoABS01@frontiersin.org Login Required This action requires you to be registered with Frontiers and logged in. To register or login click here. Abstract Info Abstract The Authors in Frontiers Fabio Babiloni Google Fabio Babiloni Google Scholar Fabio Babiloni PubMed Fabio Babiloni Related Article in Frontiers Google Scholar PubMed Abstract Close Back to top Javascript is disabled. Please enable Javascript in your browser settings in order to see all the content on this page.